Latency and ongoing replication may both contribute to the HIV reservoir, and the relative importance of each mode of persistence may vary across anatomical compartments. In the CNS, the HIV reservoir may lead to HIV Associated Neurocognitive Disorders (HAND), even in the presence of combination antiretroviral therapy (cART). We previously showed that cell-to-cell HIV spread, a directed form of intercellular HIV transmission where the virus takes advantage of cell interactions to transmit between the interacting cells, results in drug insensitivity across a range of antiretroviral drug concentratios. This insensitivity mechanism is probabilistic: since a locally high concentration of viral particle is transmitted between individual cells, the probability that at least one escapes drug action and proceeds to infect the target cell is high. Insight into the level of viral suppression achieved by cART comes from experiment-based computational models pioneered by Siliciano and co-workers. These models illustrate that the majority of currently used antiretroviral drugs (ARVs) can suppress infection far below the threshold for ongoing replication at concentrations found in the plasma. Hence, the viral particles found in the blood must be from latently infected cells that have been reactivated. Counter to this paradigm, recent data are consistent with ongoing replication in the CNS in the face of cART, including reports of relatively high viral loads, viral evolution in some patients, and immune activation. These markers for ongoing replication occur despite undetectable viral loads in the plasma. Questions remain of 1) whether ongoing replication in the CNS is feasible if current models of drug suppression are corrected for cell- to cell spread and lack of ARV penetration into the CNS, 2) whether cell-to-cell spread of CNS viral variants plays a role in persistent infection of the CNS, and 3) whether ongoing replication in the CNS is of clinical importance. In this study we will test the hypothesis that ongoing replication forms an important component of the viral reservoir in the CNS. This hypothesis will be addressed using the following study aims: 1) To determine feasibility of replication in the CNS in the face of cART. 2) To determine sensitivity of HIV cell-to-cell spread to ARVs in CNS resident cells. 3) To analyze CSF of patients on suppressive cART for viruses of macrophage/macroglial origin as markers for ongoing replication. We anticipate that this study will enable us to determine whether the CNS may support HIV transmission between cells despite suppressive cART. These results will inform control and eradication strategies for the HIV reservoir in the CNS, which are dependent on the persistence mode. Results will be especially relevant to Africa, where HIV burden is highest and where an extensive antiretroviral roll-out is taking place.